bottleneckScore 65/100Research
Karpathy identifies DevOps and infrastructure automation as unexpected constraint in building real LLM products; manually clicking through service integrations
Andrej Karpathy· Independent· AI· 2026-06-03· about Clerk, Google, Stripe, Vercel
“The code, the vibe coding part, the code was actually the easy part of vibe coding menuGen. And most of it actually was when I tried to make it real so that you can actually have authentication and payments and the domain name and the virtual deployment, this was really hard, and all of this was not code... If you try to add Google Log into your web page... it's telling me, go to this URL, click on this dropdown, choose this, go to this and click on that. It's like telling me what to do. Like a computer is telling me the actions I should be taking.”
Why it matters
Even with LLM coding assistance, deployment and service integration remains a major manual bottleneck. Developers waste significant time following click-through instructions for authentication, payments, and deployment.
Investment implication
Opportunities exist for agent-friendly APIs, infrastructure-as-code generation, and simplified service integrations (e.g., Vercel, Stripe, Clerk) that can be consumed programmatically by LLM agents. Companies automating DevOps and integration workflows for LLM apps will capture value.